07:06 <konradybcio> lumag no, i have no mpm changes

07:06 <konradybcio> what did you find?

07:37 <lumag> konradybcio, look for msm_mpm_timer_write(). I think on rb1 we are hitting the same story as we had on RPMh devices: all the cores are down and are not woken up properly.

08:05 <konradybcio> Alright I'll look into it after classes

08:38 <Marijn[m]> <Adrian[m]> "> <@marijns95:matrix.org> In..." <- Doublecheck the init sequence, pinctrl, and power, and set the brightness in prepare as I suggested before? Not getting anything on the screen, but also not getting timeouts is not something that is going to be fixed by implementing 10-bpc. I'd expect colorful corruption (or a working panel if it reads the bit depth from PPS) if bpc is wrong at least, it might be something else that got

08:38 <Marijn[m]> broken

08:51 <lumag> konradybcio, there is no pressure, I worked around it by disabling cluster sleep

09:22 <konradybcio> lumag functional cpuidle would be nice though!

09:23 <lumag> konradybcio, yep :-)

09:23 <konradybcio> Or.. I guess cpu*idle* works

09:23 <konradybcio> Just that cpunonidle doesnt

09:48 <Fekz115> Marijn[m]: I just removed checking bpc and the panel shows image correctly. For some reason it can't launch the phosh, but I see pmos splash: https://gist.github.com/fekz115/2a652ba81d6c3f910e3bd0b16d919a84

09:57 <Fekz115> And ofcourse corrupted image when I set bpc to 8

15:38 <Adrian[m]> worthing noting that the spacewar panel has `qcom,mdss-dsi-bpp = <24>;` while cupid has 30, which downstream translates to `MIPI_DSI_FMT_RGB888` and `MIPI_DSI_FMT_RGB101010` in downstream as far as I remember. I think the panels aren't as similar as it seems

15:38 <Adrian[m]> s/in downstream//

15:52 <Adrian[m]> s/worthing/worth/, s/in downstream//